Feb 26, 2013 at 2:00 pm #1299743
@maiaLocale: Rocky Mountains
Companion forum thread to:Feb 26, 2013 at 3:07 pm #1959003
"Also, as insulation is compressed, the area of contact between solid materials increases. Thus, when your insulation is 50% the thickness, your actual insulation value could be only 40%."
Citation or Reference please.
Edit: for down or synthetic batting used in garmentsFeb 26, 2013 at 3:54 pm #1959017
I believe as down compresses its insulation value per thickness unit increases, so it could be 50% compressed yet still 90% as warm (per Nisley, 2012). Other fabrics could be quite different though.Feb 26, 2013 at 4:39 pm #1959033
The thread where this was discussed http://www.backpackinglight.com/cgi-bin/backpackinglight/forums/thread_display.html?forum_thread_id=12505&startat=20
But as discussed previously to no good conclusion, the plot "The effect of Down Density, in a Fixed 3.25" baffle size, on the insulation value of sleeping bags" shows that for 16 oz of down you get 4.5 clo, 28 oz of down is 5.8 clo, so if you put 75% more down (28 – 16)/16 you get only 29% (5.8 – 4.5)/4.5 more warmth.
So, if you compress down, you lose some warmth, but only about half as bad as proportional to loft loss.
But that's just my opinion that others may disagree with.
Oh – great article – I'm going to have to read that a couple times to absorb : )Feb 26, 2013 at 7:46 pm #1959098
"This report has dealt with the degree to which different insulations are compressed by the wind. It does not discuss or attempt to measure how much insulation value (R value) is lost during compression. If there is interest in moving on to a “Part 2” on this subject, we could measure the thermal conductivities and R values of these items of clothing when compressed to different levels."
That would be interesting. Wool and fleece provide much less warmth for the weight when they're uncompressed, but their resistance to compression would partially cancel that out. It would be interesting to see numbers.Feb 27, 2013 at 9:17 am #1959230
@azajacLocale: South West
I too have great interest in seeing a part 2!Feb 27, 2013 at 9:41 am #1959238
@hknewmanLocale: Western US
Neat article but think most will shelter in high winds if they have a chance. The N3B parka was pretty warm for those of us who wore it decades ago, but the US military went to the ECWCS Gore-Tex parka and pants in the 80's and 90's, so wondering if there's a civilian alternative? Of course there's a difference between mountaineering, multi-day backcountry skiing/snowshoeing, arctic combat/construction, and UL backpacking*.
*think the UL answer is "run away" like the Monty Python skit.
Ed:*Feb 27, 2013 at 9:57 am #1959250
@dwambaughLocale: Pacific Northwest
This is interesting to note, but seems to be more of an issue for mountaineering more than backpacking. I did pay attention to the layering work-arounds that could save a hiker caught out in extreme conditions.
I have found that the air permeability of the fabric used in a lofted item is significant, having found out the hard way with a light summer sleeping bag. The fabric might as well been cheesecloth! As with adding a stiff outer shell, a bivy would have made a huge difference. In fact a very breathable shell on a sleeping bag might be considered for a bivy and bag system.
It would be interesting to see a comparison of air permeability and heat loss with an eye to moisture control. I think those are factors that would impact nearly every hiker. I would assume that this is more of an issue with synthetic fill items as down requires a fairly tight weave fabric to keep the fill at bay.
I have mentioned designing a paired windshirt and lofted fill system where the outer layer of the fill garment could be almost like cheesecloth, relying on the windshirt for the outer layer. This is in the mind that the windshirt and the outer shell of a puffy duplicate each other, adding weight and expense.
For example, I've been evaluating a Patagonia Nano Puff jacket and have come to the conclusion that the highly wind resistant shell adds as much warmth as the thin lofted insulation.Feb 27, 2013 at 10:00 am #1959252
Fortunately, I rarely see the amount of wind being discussed in this article in the Southeast, especially compared to our Western members. I do, however, thoroughly enjoy the highly technical articles (like this one) that BPL puts periodically. I hope there is a Part 2!
RyanFeb 27, 2013 at 10:28 am #1959262
@lunchandynnerLocale: Pacific Northwest
I remember there was a thread here about thermolite vs prima loft.
Somebody had said that Thermolite did not lose as much warmth when compressed by wind which is why it is used in high altitude jump/skydiving suits and what not. Anybody know if this is true?Feb 27, 2013 at 11:11 am #1959285
@richard295Locale: San Francisco Bay Area
See thread http://www.backpackinglight.com/cgi-bin/backpackinglight/forums/thread_display.html?forum_thread_id=42340 and my post on
02/03/2011 14:04:58 MSTFeb 27, 2013 at 12:53 pm #1959335
Great article. However, I have a hard time imagining that applying a stack of weights onto fabric in a horizontal plane in a static environment represents what happens under wind compression conditions. Yes, there are winds that blast from mostly one direction in a sustained manner, but most of the time, even in a sustained wind, the wind "gusts" and is multidirectional. You did mention how wind "pumps" the heat out of loft, so I'm sure you can see that the static pressure plate model has its limitations.
I too would like to see a Part 2. For a future article, I'd like to see someone explore what else is happening when the wind strikes a warm being in a jacket. For example, the air that is deflected around the sides of the person would also cause some compression but might there also be a vacuum effect drawing further heat away? And what impact would the texture of the outer fabric layer have on heat loss?
And on the subject of wool, I suspect the lower compression was due mostly to the knit or weave. Makes me wonder if some of those bumpy knit patterns worn by Irish fishermen aren't just ornamentation but serve to trap more air and prevent compression when worn under a rain parka? I also wonder if lanolin left in the wool makes a difference in heat loss from wind. Wool fleece, which is not woven, would seem more likely to felt up after awhile and completely lose its loft. On the other hand, felted wool traps a lot of warm air and is fairly stiff so it could make a good outer layer under a water-resistant layer. MaxineFeb 27, 2013 at 2:35 pm #1959393
Bogs and BergsMember
"Makes me wonder if some of those bumpy knit patterns worn by Irish fishermen aren't just ornamentation but serve to trap more air and prevent compression when worn under a rain parka?"
Maxine, I don't have any science to quote, but as a knitter in a centuries-old knitting culture, I'd say you're spot on. If a person needs a Serious Work Sweater, the cable needles soon come out, and it's not out of need for ornamentation. Air pockets are warmer than flat knitting, and the material becomes much 'thicker' without using thicker wool (less flexibility/mobility) or decreasing its density (by using larger needles).
I bet you're also right that "lanolin left in the wool makes a difference in heat loss from wind". The fibre will be denser and have a slicker, almost shiny surface, so more wind slips over rather than through the fabric. The lanolin is also a natural DWR, and especially with felting, it's possible to turn high-lanolin wool into a nearly waterproof material. On a well-felted mitten, water will bead and roll off even under a good amount of pressure.Feb 27, 2013 at 2:40 pm #1959395
@stephen-mLocale: Way up North
Those Irish Fisherman Sweaters are called Aran Jumpers.
My mother used knit them many moons ago.Feb 27, 2013 at 2:58 pm #1959401
I agree that our methodology assumed a steady state wind condition that would not represent gusty winds. However, it represents the peak of wind gusts and some people who have been in very strong winds report that they are not always gusty. To evaluate wind gusts and the real anatomy of a human would probably require a manikin and wind tunnel. This was beyond my budget for this article. Maybe a topic for future study.
Yes, the air deflected around would cause a different result on the sides and back. It is possible that a vacuum effect on the rear due to a vortex might actually increase the loft slightly. I doubt that it would make a person warmer on that side when gusty.
I agree that the knit of wool can be very important in trapping dead air and breathability. Irish fishermen should know a thing or two about keeping warm in a wet, windy environment. The lofted wool however was not knit as far as I know.Feb 27, 2013 at 5:44 pm #1959473
Add me to the list of folks who would love to see a followup. This is the kind of stuff that really separates this site from the crowd. Anyone can review a piece of gear, but you dig into the science of it all, and I love it.
It wouldn't surprise me if down is still the best material, ounce for ounce, even in heavy winds. As folks have suggested, it is possible that its heat loss is not directly proportional to its loss in loft. It is also possible that even with a higher proportion of loss, it is a better insulator (ounce for ounce) than anything else. In other words, if down starts out being three times as good as the competition, it may still be a better material if it is only twice as good once the wind starts blowing.Feb 27, 2013 at 5:49 pm #1959476
– -K.T.- –Participant
I can see this being of use to skiers. But how does this transfer to backpackers wearing a pack? A puffy and a pack on would make for some wildly different thicknesses of insulation.Feb 27, 2013 at 9:11 pm #1959564
I would like to add a comment for clarity to the article.
All insulation is composed of solid materials that are touching each other that are conductors of heat (such as down fibers). Between these solid materials there are dead air pockets that resist the flow of heat. As insulation is compressed, the ratio of area of the solid materials (conductors) that are touching each other increases. In addition, the area of air pockets decreases. This reduces the insulation value ("R" value).
I have never encountered any exceptions to this. All materials decrease in insulating efficiency as they are compressed. Each material decreases at a different rate.
A part 2 study will demonstrate this more clearly.Feb 27, 2013 at 10:20 pm #1959579
@richard295Locale: San Francisco Bay Area
Your explanation correctly states that fiber conductivity increases with increased density but, omits the offsetting reduced radiation losses over a significant density variance.
A forum search on related terms will point you to both primary and secondary related research that I won't bother to repeat.Feb 28, 2013 at 12:57 pm #1959764
It was my intent to keep part 1 fairly simple and basic and not overwhelm the first readers.
You all make good points. Thanks. Please keep them coming so I might be able to address them in a part 2 article. Heat transfer is a complicated science and there are many issues to consider.
For instance, the "bellows effect" or pumping of the warm air out of the insulation is another issue that should be considered. I have often felt that this might be the largest method of heat loss in gusty winds. Certain storm shells could reduce this.
All these issues lead up to the reason that more testing should probably be done. If so, I will try to address some of the other issues in that report.Mar 1, 2013 at 9:49 pm #1960363
As you compress down, there is an increase in conduction from the materials contacting each other, but there is also an increase in the number of air pockets. This results in a substantial increase in the insulation value per thickness unit, but still a small decrease in overall efficiency (ie. fully lofted is best, but not by much).
When you say a down garment compressed to 50% might actually by only 40% as warm, you're double counting the same effect by saying it's thinner plus less warm (ie. 50% thick x 80% insulating = 40%). Actually, it's thinner but much warmer per thickness unit (ie. 50% x 180% = 90%). so it can be compressed by 50% yet only decreased in warmth by 10%.
Anyways, Will explains this much better than I in BPL's down jacket state of the market:
And I agree that in truly windy conditions the bellows effect is probably a bigger deal anyways.Mar 2, 2013 at 5:24 am #1960412
While they were not involved with clothing, it might be useful to look at the R-value vs thickness results in Roger's and Will's 2011 tests and articles about many many insulated sleeping mats.
They definitely found a decrease in R-value as mats were compressed (pads compressed from full inflation thickness to quite thin in 5mm increments)Mar 2, 2013 at 7:52 am #1960457
"When you say a down garment compressed to 50% might actually by only 40% as warm, you're double counting the same effect by saying it's thinner plus less warm (ie. 50% thick x 80% insulating = 40%). Actually, it's thinner but much warmer per thickness unit (ie. 50% x 180% = 90%). so it can be compressed by 50% yet only decreased in warmth by 10%."
The insulation is provided by air. The conductivity of the fibers is not a factor because they are a small part of the volume.
The problem is that there is convection inside the insulation. Air moves around carrying it's heat with it. The fibers of the insulation reduce that convection.
What's good about down is that the fibers are very fine and light weight. But the downside is they impede internal convection less than, for example, synthetics. My measurements:
clo/oz/yd2 – down = 1.29, apex = 0.81, apex is 0.63 of down
clo/inch – down = 2.28, apex = 3.2, apex is 1.4 of down, theoretically, air is 6.4 clo/inch without internal convection or radiation
If you compress down, the thickness of the air layer is less so there's more conduction heat loss. This is partially cancelled because the internal fibers are closer together, which impede internal convection better. My measurements: when I compressed the down 47% (e.g. the loft reduced from 1 inch to 0.53 inch) the clo/oz/yd2 was reduced by 26%, and the clo/inch increased by 39%. If there was no change in internal convection, then the clo/oz/yd2 would be reduced 47%.
Measuring this is tricky so you should take my numbers with a grain of salt, but that table of Richard's is consistent. I haven't seen anyone else measure this which is why I did.
This all ignores radiation which is rather non-intutive to me, so I don't know. Richard says radiation is why the compression of down doesn't reduce the insulation value. I don't know. It seems like that would show up in Richard and my measurements…
The only reason to talk about this, is, if you're making a down garment, make the baffles big enough so it's not over-stuffed by more than about 20 or 30% if you want to get the most warmth for the weight. But, even if you overstuffed by 100% or 150%, you would hardly notice. Maybe you'de need 1 pound of down compressed 20% or 30%, 1.5 pounds if it was compressed 150% to get the same warmth.
And anecdotally, they say to fluff the down up good when you unpack to get the most warmth. If the baffles aren't big enough, then they won't allow the down to do this.Mar 2, 2013 at 8:07 am #1960465
Jim, yes virtually everything decreases in R-value as you compress it. However, there's a key distinction here between the total insulating ability (ie. R-value) and the insulating ability per thickness unit, which is what I'm trying to communicate.
Down is a better example, but lets look at Roger's NeoAir test data (2009). He found:
Pad at 2.5" thick = R-3 (R-value per inch = 1.2)
Pad at 2.0" thick = R-2.5 (R-value per inch = 1.3)
So we see a decrease in total insulating ability as the pad gets thinner, yet the pad is actually getting warmer per inch – there's just less inches. So the main point is that thickness is down by 20%, but R-value is only down by 13%.
A similar thing is true for down (and virtually any insulation). As they are compressed they get warmer per inch, but there's less inches so the overall result is a loss. With down this loss is really small. It can be 50% as thick yet still have total R-value at ~90%. This seems to be what Jerry is saying as well.Mar 2, 2013 at 8:25 am #1960474
Yes, that's what I'm saying
Except – with down, if you compress 50%, it's only about 25% less warm
So it's significant, but only half as bad as it could be
But measuring and interpreting is tricky so maybe I'm wrong
You must be logged in to reply to this topic.